Price-Making Storage Strategy

Updated 4 February 2026

Price-making storage strategies are approaches where energy storage optimizes its operation by internalizing its impact on market prices, leading to strategic withholding and modified bidding behaviors.
The methodology utilizes self-scheduling models and Cournot-type competition with KKT conditions to incorporate linear endogenous price responses in decision-making.
These strategies affect market power and welfare by calibrating price sensitivity, implementing bid caps, and providing detection rules to distinguish strategic from competitive behavior.

A price-making storage strategy refers to the explicit optimization, operational, and detection methodologies employed by energy storage units (ESUs) that anticipate, internalize, and exploit their endogenous impact on electricity market clearing prices. Unlike price-taking models, where storage assumes no price impact, price-making approaches capture the feedback loop between strategic storage scheduling and market prices, often leading to capacity withholding, non-smooth operating profiles, and modified welfare and profit outcomes. Such strategies are especially relevant in modern electricity markets characterized by high penetrations of flexible storage, restricted conventional generator flexibility, and volatile or inelastic net load profiles.

1. Price-Making Self-Scheduling Models

A core foundation for price-making storage is the self-scheduling optimization that incorporates price impact. For an energy-constrained storage unit operating over discrete intervals $t\in\mathcal{T}=\{1,\dots,T\}$ , denote $p_t$ (discharge), $b_t$ (charge), $e_t$ (state-of-charge), with technical limits $\bar P$ , $E$ , round-trip efficiency $\eta$ , and a price sensitivity $\alpha_t$ dictating the linear price response to net injection.

The price-taker model solves: $\max_{\{p_t,b_t,e_t\}} \sum_{t=1}^T \hat{\lambda}_t (p_t-b_t)$ subject to operational and SoC constraints, using exogenous price forecasts $\hat{\lambda}_t$ (Wu et al., 2024).

The price-maker model prescribes linear endogenous price response: $p_t$ 0 and maximizes: $p_t$ 1 yielding a convex QP whose KKT system couples storage decisions to price impact. In Cournot-type models, as in balancing or fully renewable markets, firms submit quantities (charging/discharging levels) anticipating market feedback induced by their bids (Abate et al., 30 Sep 2025, Cruise et al., 2016).

2. Market Power, Withholding, and Optimal Bidding Rules

Strategic ESUs with market power exploit the tradeoff between immediate profit and systemic price impact by capacity or economic withholding. The KKT conditions for the convex QP yield optimality relations: $p_t$ 2

$p_t$ 3

with SoC coupling and complementarity. For two-period problems, the analytic rule exhibits switching thresholds: full-capacity operations occur when price-differential exceeds a function of $p_t$ 4, else partial operation arises. Intuitively, storage can maximize spread capture either by shifting maximally between two price extrema, or—if the price impact penalty is too large—by partial cycling that modulates market prices (Wu et al., 2024).

In Cournot competition among $p_t$ 5 symmetric stores under $p_t$ 6, each firm in equilibrium supplies $p_t$ 7. As $p_t$ 8 increases, individual volumes and profit decay sharply, restoring social welfare but suppressing rents—core to the self-limiting nature of storage market power (Cruise et al., 2016, Abate et al., 30 Sep 2025).

3. Detection and Characterization of Withholding in Storage Operation

Market operators require robust ex-post criteria to screen observed schedules for price-making behavior. Theorem 1 in (Wu et al., 2024) establishes a sufficient test: count $p_t$ 9 partial-capacity intervals where $b_t$ 0 or $b_t$ 1; if $b_t$ 2, market power is being exercised. Even if $b_t$ 3, one must check that KKT-implied price-quantity equalities for competitive behavior hold; deviations signal strategic withholding. These rules are implementable via profile monitoring without perfect knowledge of storage forecasts or market fundamentals.

4. Price Sensitivity, Demand Elasticity, and Systemic Effects

The degree of price impact $b_t$ 4—often empirically calibrated as the negative slope of the residual demand curve—quantifies the potential for market power. High $b_t$ 5 settings enable greater profits from price-making: for example, a 2.5 MW/10 MWh battery, under winter NYISO prices with $b_t$ 6/\text{MW}^{2\cdot\text{h} $,</sup> raised profit by 30% via withholding near price peaks (<a href="/papers/2405.01442" title="" rel="nofollow" data-turbo="false" class="assistant-link" x-data x-tooltip.raw="">Wu et al., 2024</a>).</p> <p>Market design and regulatory structures greatly influence outcomes:</p> <ul> <li>Introduction of moderate short-run demand elasticity ($ b_t$7) smooths prices, dramatically reducing zero-price hours (from 90% to ≈30%) and volatility, thereby muting potential for extreme storage rents (Brown et al., 2024).}

In oligopolistic settings, increasing the number of competing storages rapidly erodes price distortions and market power: as few as 3–4 active players restore near-social-optimal prices/welfare (Abate et al., 30 Sep 2025, Cruise et al., 2016).

Price caps or explicit limiting of bid bands can stabilize games otherwise prone to non-convergent, infinite undercutting by strategic storages when conventional flexibility is low (Navon et al., 2024).

5. Extension to Supply Function Equilibrium and Multi-Market Settings

The price-making paradigm generalizes to other market architectures:

Under supply function equilibrium (SFE), storages submit affine supply functions, anticipating that increases in their net injection lower prices linearly (via $b_t$8). The resulting equilibrium is a function of all participant strategies and the residual demand slope $b_t$9. The joint KKT system defines the SFE (Wu et al., 2024).
In multi-product, multi-market clearing (energy, reserve, regulation), price-making storage strategies exploit cross-market couplings, adjusting bids for energy versus ancillary services within an MILP or bilevel optimization. Withholding in high-margin products (e.g., regulation mileage) can shift relative profitability and affect system prices, as observed empirically in joint market clearing (Khalilisenobari et al., 2020).

6. Welfare Effects, System Costs, and Policy Implications

While price-making storage can increase rents for strategic units, its systemic impact is generally welfare positive up to a limit. Withholding in anticipation of more volatile or uncertain prices can reduce total system cost when storage hedges real price risks optimally. However, absent constraints, extreme price volatility (unbounded forecast variance) admits unbounded bids, mandating sensible volatility limits and bid caps (Qin et al., 2024). Notably, storage economic withholding due to opportunity cost does not equate to manipulative market power; rather, it reflects rational profit-maximization under multi-period uncertainty.

From a regulatory perspective:

Moderate storage penetration and competition support both market stability and cost-efficiency.
Policy levers—price or bid caps, mandatory reporting of capacity-specific schedules, or demand participation—can be calibrated to balance incentives for flexible arbitrage against the risks of concentrated market power.
Ongoing ex-post screening based on high-frequency operation profiles and the sufficiency criteria described above remain essential for market resilience (Wu et al., 2024, Abate et al., 30 Sep 2025).

7. Implementation Considerations and Future Directions

Optimal price-making storage strategies require:

Realistic calibration of price sensitivity $e_t$ 0, SoC-dependent bidding trajectories, and accurate forecasts of rival behaviors.
Dynamic programming or rolling-horizon solutions to capture intertemporal tradeoffs, with extensions to stochastic settings via SDP or reinforcement learning to navigate non-stationary environments (Badoual et al., 2021).
Periodic re-optimization as system fundamentals evolve: higher renewable penetration, increased storage deployment, or regulatory and market design changes reshaping value streams and profit opportunities.

Continued research aims to quantify the bounds of strategic withholding under increasing market complexity, extend ex-post detection approaches, and develop real-time market power warnings based on observed scheduling and supply function deviations.

Key sources supporting these results and methods include (Wu et al., 2024, Abate et al., 30 Sep 2025, Cruise et al., 2016, Qin et al., 2024, Navon et al., 2024, Brown et al., 2024, Khalilisenobari et al., 2020).